10/20/2014

711.11 watts 355.55 watts 127.90 watts 102.51 watts 70.41 watts

All these numbers from one little old Super 16 D. Who would have thought?

That first number is the peak in rush wattage produced by a .360 resistance at 16 volts.

The 355.55 watt value is the mean average input over the full rpm of the motor. The value with the

motor reaching just above 1/2 of it's no load speed, roughly. Remember that as the motor spins a back

voltage opposes the forward voltage. The motor then sees full voltage at stall and just enough voltage

to overcome the motors "losses". That's the current you see on your power supply when voltage is

steady and the load absent. These are your I2R or copper losses and friction.

127.90 watts represents the peak electrical power being utilized of those 355.55 watts in an

attempt to make usable horsepower at the pinion. 36% efficiency but electrical efficiency alone doesn't

make horsepower whatsoever. Armature iron and magnets for the sake of simplicity turn that "juice" into

torque. Another number you hear tossed about. Torque density. How many units of torque per amp of

juice.

The fourth value, 102.51 watts is the value of the actual power at the pinion at peak measured

electrical efficiency. This is the number a dyno spits out. Pretty useful number but it can also be

calculated with decent accuracy. Here is you bragging efficiency number. 29% for this motor. The

percent of shaft power to input power and this is the classic definition of motor efficiency in the

textbooks. Power in verses power out.

The last number is RMS wattage. 70.41 watts. That value the motor delivers over the entire rpm

range. This is the actual number that moves the car. There is another number and that's the area

under the curve in percent. This motor reported 69%.

The last two paragraphs are the topic for just a bit longer. This is the area I have been working on

for years and writing about so I included them for the sake of a complete story. The program goal for

quite some time was to do what others traditionally have done then go a step further. Increase electrical

efficiency. Then increase the area under the curve but there are issues with that plan as good as it is

even though it has produced some killer power with average and mixed manufacture components.

First is that the dyno reads the values directly such as current plus RPM and delta T or time

differential and calculates the outcome. Part of that calculation, actual power under load output is a

direct measurement. So is RPM and RPM to time increments. On the input side there is a fly in the

soup. It calculates the resistance based on the voltage regulators target value and direct reading of

current. Unless your really paying attention to your dyno reports and not looking at the pretty pictures it

draws you it can give a head fake. Many report values are calculated for the assumed input voltage.

Such as power factor and motor efficiency per unit time. When reported peak or average efficiency

goes up you better give input power factor a look before you let turn out the lights in the motor room.

So what more can be done? Not much without some better tools. Like the chart above displays.

Quantification of the efficiency between the electrical efficiency and the output efficiency. Your

thinking that is just a different way to say overall motor efficiency. Not at all. This is a comparison

between the actual measured electrical successfully delivered and shaft power. For this motor that

difference is plotted in the chart. Believe it or not on every point on this chart the lower line is exactly

80.15% below the upper line.

So, where is the difference going? Lets start with commutation. Armature resistance was a

measured .360 Ohms. A four wire Kelvin rotating armature terminal resistance reading gave .5002

Ohms. A 72% difference in available power. 72% of 80% is 90%. That percentage is the loss from

commutation errors. Work that down the chain and it represent 18 pinion watts of power added to the

102 watts currently present. 120 watts total. Where is the other 10% going? Later.

What we do know is the both lines in the chart move in lock step and correcting this moves the

Peak power input up 200 watts. We will never eliminate I2R losses meaning there are 100 electrical

watts left and how many of those we can capture in based in our skills at reducing the resistance level.

Here is an example. This method pinpoints the areas that need work. What the problems are and

hints at their correction.

The other 10% or 22 additional watts are pure iron issues. Parma steel. We have now accounted

for losses on the electrical side. Copper losses, commutation losses and iron losses.

1.) Copper losses 50% of Peak input, or 200 watts based on resistance difference and voltage.

2.) Iron losses 10% of this difference is iron losses

3.) 90% of this is additional commutation resistance.

Moving ahead we get into the top chart. What prevents the electrical power from becoming like

shaft power. A 20% difference! It isn't armature resistance. It is how effective the iron flux and magnet

flux are at converting current into torque. Lets segment groupings.

Group 1 is getting all the current your winding is capable of. A combination of resistance, stray iron

losses and inductance. Both iron losses in the form of eddy currents and inductance are subtractions

from voltage and resistance. They are in fact additional resistance.

Group 2 is converting all that current into as much torque as possible while reducing mechanical

losses and limiting torque ripple.

Group 3 is maximizing the area under the curve. Enhancing the RMS of the peak.

So, we need a litmus test for each group to test with.

Group 1 is as simple as a 4 wire Kelvin measurement on the rotating assembly without magnets in

the motor.

Group 2 is a bit harder but still doable. Do a generator test and measure the voltage noting the

rpm. Covert this information into RPM per 1 volt. Next do the motoring test. Drive the motor at around 4

or 5 volts just long enough to take an RPM reading. Convert that to RPM per 1 volt and divide this

number by the first. That is the difference between the theory and the actual conditions we represented

in the chart. It's also the difference between forward voltage and BEMF. Last do what you can to limit

the no load amp draw at four volts or so that is caused by friction and or windage losses.

Group 3 is a calculation if ballpark is good for you OR actual dyno data work ups if you need more

precision. This boys is the area the winners are separated from the also ran. That the rough out.

I'll cover BEMF which is part of group 2 in a future posting. It needs it's own room.

711.11 watts 355.55 watts 127.90 watts 102.51 watts 70.41 watts

All these numbers from one little old Super 16 D. Who would have thought?

That first number is the peak in rush wattage produced by a .360 resistance at 16 volts.

The 355.55 watt value is the mean average input over the full rpm of the motor. The value with the

motor reaching just above 1/2 of it's no load speed, roughly. Remember that as the motor spins a back

voltage opposes the forward voltage. The motor then sees full voltage at stall and just enough voltage

to overcome the motors "losses". That's the current you see on your power supply when voltage is

steady and the load absent. These are your I2R or copper losses and friction.

127.90 watts represents the peak electrical power being utilized of those 355.55 watts in an

attempt to make usable horsepower at the pinion. 36% efficiency but electrical efficiency alone doesn't

make horsepower whatsoever. Armature iron and magnets for the sake of simplicity turn that "juice" into

torque. Another number you hear tossed about. Torque density. How many units of torque per amp of

juice.

The fourth value, 102.51 watts is the value of the actual power at the pinion at peak measured

electrical efficiency. This is the number a dyno spits out. Pretty useful number but it can also be

calculated with decent accuracy. Here is you bragging efficiency number. 29% for this motor. The

percent of shaft power to input power and this is the classic definition of motor efficiency in the

textbooks. Power in verses power out.

The last number is RMS wattage. 70.41 watts. That value the motor delivers over the entire rpm

range. This is the actual number that moves the car. There is another number and that's the area

under the curve in percent. This motor reported 69%.

The last two paragraphs are the topic for just a bit longer. This is the area I have been working on

for years and writing about so I included them for the sake of a complete story. The program goal for

quite some time was to do what others traditionally have done then go a step further. Increase electrical

efficiency. Then increase the area under the curve but there are issues with that plan as good as it is

even though it has produced some killer power with average and mixed manufacture components.

First is that the dyno reads the values directly such as current plus RPM and delta T or time

differential and calculates the outcome. Part of that calculation, actual power under load output is a

direct measurement. So is RPM and RPM to time increments. On the input side there is a fly in the

soup. It calculates the resistance based on the voltage regulators target value and direct reading of

current. Unless your really paying attention to your dyno reports and not looking at the pretty pictures it

draws you it can give a head fake. Many report values are calculated for the assumed input voltage.

Such as power factor and motor efficiency per unit time. When reported peak or average efficiency

goes up you better give input power factor a look before you let turn out the lights in the motor room.

So what more can be done? Not much without some better tools. Like the chart above displays.

Quantification of the efficiency between the electrical efficiency and the output efficiency. Your

thinking that is just a different way to say overall motor efficiency. Not at all. This is a comparison

between the actual measured electrical successfully delivered and shaft power. For this motor that

difference is plotted in the chart. Believe it or not on every point on this chart the lower line is exactly

80.15% below the upper line.

So, where is the difference going? Lets start with commutation. Armature resistance was a

measured .360 Ohms. A four wire Kelvin rotating armature terminal resistance reading gave .5002

Ohms. A 72% difference in available power. 72% of 80% is 90%. That percentage is the loss from

commutation errors. Work that down the chain and it represent 18 pinion watts of power added to the

102 watts currently present. 120 watts total. Where is the other 10% going? Later.

What we do know is the both lines in the chart move in lock step and correcting this moves the

Peak power input up 200 watts. We will never eliminate I2R losses meaning there are 100 electrical

watts left and how many of those we can capture in based in our skills at reducing the resistance level.

Here is an example. This method pinpoints the areas that need work. What the problems are and

hints at their correction.

The other 10% or 22 additional watts are pure iron issues. Parma steel. We have now accounted

for losses on the electrical side. Copper losses, commutation losses and iron losses.

1.) Copper losses 50% of Peak input, or 200 watts based on resistance difference and voltage.

2.) Iron losses 10% of this difference is iron losses

3.) 90% of this is additional commutation resistance.

Moving ahead we get into the top chart. What prevents the electrical power from becoming like

shaft power. A 20% difference! It isn't armature resistance. It is how effective the iron flux and magnet

flux are at converting current into torque. Lets segment groupings.

Group 1 is getting all the current your winding is capable of. A combination of resistance, stray iron

losses and inductance. Both iron losses in the form of eddy currents and inductance are subtractions

from voltage and resistance. They are in fact additional resistance.

Group 2 is converting all that current into as much torque as possible while reducing mechanical

losses and limiting torque ripple.

Group 3 is maximizing the area under the curve. Enhancing the RMS of the peak.

So, we need a litmus test for each group to test with.

Group 1 is as simple as a 4 wire Kelvin measurement on the rotating assembly without magnets in

the motor.

Group 2 is a bit harder but still doable. Do a generator test and measure the voltage noting the

rpm. Covert this information into RPM per 1 volt. Next do the motoring test. Drive the motor at around 4

or 5 volts just long enough to take an RPM reading. Convert that to RPM per 1 volt and divide this

number by the first. That is the difference between the theory and the actual conditions we represented

in the chart. It's also the difference between forward voltage and BEMF. Last do what you can to limit

the no load amp draw at four volts or so that is caused by friction and or windage losses.

Group 3 is a calculation if ballpark is good for you OR actual dyno data work ups if you need more

precision. This boys is the area the winners are separated from the also ran. That the rough out.

I'll cover BEMF which is part of group 2 in a future posting. It needs it's own room.

Breaking the Code 1 |